Daisy Intelligence

Gary Sarenvarada left his position running IBM Canada's data mining practice in 2003, shocked by how little math and science informed corporate decision-making. With a background in aerospace engineering and expertise in neural networks and machine learning from the 1990s, he saw an opportunity to apply sophisticated mathematical models to business problems. His vision was audacious: solve problems beyond human capability—highly repetitive, data-intensive decisions made millions of times daily. But the mission went deeper: if he could make retailers more efficient, they'd lower prices, reducing the cost of living for consumers. He founded Daisy Intelligence to fulfill what he calls "the promise of the information age."

Gary initially bootstrapped the company alongside a professional services business that helped build the core IP. For over a decade, Daisy remained under the radar, perfecting its machine learning engine. The turning point came in 2016 when the company shifted to 100% recurring revenue. Gary raised capital from super angels and secured venture debt from Espresso Capital in Toronto, totaling $4.5 million to date. The tech stack relies on reinforcement learning—a technique Gary claims no other company uses outside of engineering—which delivers outsized value by making autonomous recommendations on pricing, promotions, and inventory allocation.

Daisy targets enterprise retailers at the C-level, focusing on companies with $100 million to $10+ billion in annual revenue. Customers pay between $250,000 and $500,000 annually (averaging $20,000/month). Gary's go-to-market is consultative and high-touch: he sells directly to executives who make quick decisions when they understand the ROI. The company has 17 customers globally—in Canada, the US, New Zealand, and Europe—with a sales cycle of 6-12 months. The sales process is efficient because Gary pitches at the C-level where executives rapidly grasp whether Daisy's vision aligns with their needs.

Daisy's value prop is compelling: customers who execute its recommendations see their net income double—a claim Gary stands behind with a guarantee. "If we don't deliver 10 times the return, we tell our customers we're gonna quit and move on," he says. The metrics prove it works. The company has achieved 110% net revenue retention with -10% revenue churn (losing a few customers but expanding others by 20%) and negative gross churn overall. Customer acquisition cost is 2.5x annual contract value, translating to a four-month payback period.

The main friction: change management. Some merchants resist an AI system "stepping on their toes" around pricing and promotion decisions—their traditional domain. This caused early customer losses, but the companies that embraced the technology expanded their contracts by adding additional modules in year two.

Daisy is at a $4 million ARR run rate, having doubled revenue in the past year (100% YoY growth). With a team of 40 split between Toronto and Ukraine, Gary is fundraising $10 million at a $40 million pre-money valuation to accelerate expansion. His roadmap is aggressive: scale to one new global market with the Series A, then pursue Series B to enter Europe, Asia, and Latin America. He envisions 50-60 customers within three years—a unicorn valuation of nearly $1 billion. At 52 years old, Gary regrets not starting earlier, but Daisy Intelligence is now executing at a scale that proves his thesis: machine intelligence, properly deployed, can transform how enterprises operate.